Propensity Models for Detecting Misclassified Merchant Codes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in accurately identifying and correcting misclassified merchant category codes (MCCs) assigned to merchants, which can lead to incorrect risk assessment, interchange fee determination, and reward distribution, often requiring manual and resource-intensive investigations.
Innovation Solution
A machine learning-based system using propensity models analyzes consumer transaction data to automatically identify and correct misclassified MCCs by comparing merchant behavior with expected patterns, without manual review or additional data collection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual investigation methods are used to verify MCC classification accuracy, then measurement precision can be improved, but productivity deteriorates due to time-consuming and resource-intensive processes
Solution Approach 1:
The patent replaces manual mechanical investigation processes with an automated machine learning system. The ML model analyzes transaction data patterns to automatically detect misclassified MCCs, eliminating the need for manual review while maintaining or improving detection accuracy. This substitution enables high-volume processing without proportionally increasing human resources.
Solution Approach 2:
The patent introduces an intermediary ML-based detection system between the MCC assignment process and final classification verification. This intermediary automatically analyzes transaction patterns and flags potential misclassifications, serving as a mediator that reduces the burden on manual verification processes while improving overall system accuracy.
2Measurement precision
If additional computing resources and personnel are deployed to investigate merchants one-by-one, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent creates a universal ML-based detection system that can handle multiple acquiring banks and numerous merchants simultaneously through a single centralized platform. This multi-functional system eliminates the need for each acquiring bank to deploy separate investigation teams and infrastructure, reducing overall system complexity while improving detection precision across the entire network.
Solution Approach 2:
The patent merges the MCC verification functionality into the existing payment processing network infrastructure. By combining detection capabilities with the central network that already processes all transactions, the system avoids creating separate complex verification infrastructures. The ML model leverages existing transaction data flows to perform detection without requiring duplicate systems.
3Productivity
If propensity models are trained and deployed for automatic MCC verification, then productivity improves through automation, but device complexity increases due to machine learning infrastructure requirements
Solution Approach 1:
The patent performs preliminary action by training the propensity models in advance using historical transaction data with known correct MCC classifications. This pre-training phase creates ready-to-use detection models that can be deployed without requiring complex real-time training infrastructure. The models are prepared beforehand to handle production workloads, simplifying the deployment architecture while maintaining high productivity.
Data Source
AI summary
A computer system and method having a machine learning tool for identifying and correcting a misclassified merchant category code (MCC). The system includes a computer device that has at least one processor configured to store a first propensity model that is trained with multiple account identifiers that are used to initiate multiple purchase transactions with multiple merchants each having been properly assigned to a first MCC. The system inputs into the first propensity model an account identifier used to initiate a purchase transaction with a candidate merchant assigned to the first MCC. The candidate merchant possibly being mis-assigned to the wrong MCC. The system outputs from the first propensity model a first score based on the inputted account identifier, compares the outputted score to a threshold value, and based on the comparison, determines that the candidate merchant was mis-assigned to the first MCC.